Almost every quality article on this site ends the same way: a matching intact mass confirms composition, not sequence order — for that you need MS/MS. That sentence has now appeared here more than a dozen times as a deferral. This article is the thing being deferred to.
Tandem mass spectrometry and peptide mapping are the two techniques that actually address the question "are the amino acids in the order the label claims?" They are also routinely named on documentation in ways that do not match what the technique can do for the molecule in question. Understanding the mechanics is what separates a meaningful identity claim from a line item.
What a matching intact mass leaves open
A conventional identity package pairs HPLC purity with an intact mass measurement. The mass spectrometer reports the molecular weight of the whole molecule, and if it matches the theoretical mass of the intended sequence, the compound is composed of the right atoms.
That is a real result, and it excludes a great deal. It does not exclude:
- A scrambled sequence. Reorder every residue and the elemental composition is unchanged. The mass is identical.
- An epimer. D- and L-amino acids are isobaric.
- A mis-paired disulfide. Connectivity isomers weigh the same.
These are not exotic failure modes. They are the specific things a synthesis route can produce, and they are the reason a second, structural measurement exists at all.
The sequence ladder: how MS/MS reads order
In tandem MS, a selected precursor ion is fragmented inside the instrument and the fragments are mass-analyzed in a second stage. Peptides break preferentially along the backbone, and the resulting pieces fall into named series depending on which terminus retains the charge. The nomenclature comes from Roepstorff and Fohlman's 1984 proposal in Biomedical Mass Spectrometry, later extended by Biemann to cover side-chain cleavages: N-terminal fragments are a, b, c; C-terminal fragments are x, y, z, with a subscript for the number of residues.
The useful consequence is arithmetic. The mass difference between consecutive ions in a series equals the residue mass of the amino acid between them. Line up the b-ion series and read the gaps and you have read the sequence left to right; do the same with the y-series and you read it right to left. The two ladders overlap and cross-check each other.
This is what "MS/MS confirms sequence" means concretely. It is not a fingerprint comparison — it is a residue-by-residue readout, and its quality is measured by how much of the backbone the observed fragments actually cover.
Peptide mapping is a protein test, and most of this shelf isn't a protein
Here is where documentation and chemistry frequently diverge.
Peptide mapping is a defined analytical procedure: digest the molecule with a sequence-specific protease, separate the resulting fragments chromatographically, and compare the pattern — usually with MS on the peaks — against a reference. USP General Chapter ⟨1055⟩ Biotechnology-Derived Articles—Peptide Mapping is the formal treatment, harmonized with the corresponding JP and EP chapters. Its revised version became official December 1, 2024, and the revision is telling: the Introduction was rewritten to define peptide mapping as a chemical identification test, and the chapter's scope was narrowed to confirmation of identity. Sections on experimental aspects and data comparison were deleted; a "Points to Consider Prior to Validation" section was added.
Note the framing in the chapter's own title. It is a test for biotechnology-derived articles — recombinant proteins large enough that you cannot fragment them intact and must cut them into analyzable pieces first.
Most compounds on a research-peptide shelf are not in that category, and for many of them the procedure does not physically work. Trypsin — the default mapping enzyme — cleaves C-terminal to lysine and arginine, but the long-standing Keil rule holds that Lys-Pro and Arg-Pro bonds are resistant. Apply that to the catalog:
- BPC-157 (GEPPPGKPADDAGLV) has one lysine, at position 7, immediately followed by proline. No arginine anywhere. Trypsin has nothing it can cut.
- Selank (TKPRPGP) has a lysine followed by proline and an arginine followed by proline. Same result.
For a 15-mer or a 7-mer, digestion is not merely unnecessary — it is the wrong operation. A short peptide is small enough to be fragmented intact in the instrument, producing the full b/y ladder in one experiment. That is genuine MS/MS sequence confirmation, and it is more informative than any digest would be.
So a COA reading "peptide mapping" against a heptapeptide is doing what an amino acid analysis line on a small molecule does: reproducing a template. Conversely, for IGF-1 LR3 at 83 residues or HGH 191AA at 191, peptide mapping is exactly the right test and its absence is a real gap. The same phrase means different things at different lengths — a theme that runs through manufacturing route and impurity profile as well.
Where the ladder goes quiet
An MS/MS spectrum is not a clean readout of every bond. Two limitations matter.
Fragmentation is uneven. Under collision-induced dissociation, cleavage N-terminal to proline is strongly favored — the well-documented "proline effect" — and a dominant cleavage can suppress the rest of the ladder. This shelf is unusually proline-heavy: BPC-157 carries three consecutive prolines plus two more, and the Pro-Gly-Pro chassis shared by Selank and Semax is proline by design. Uneven fragmentation is why sequence coverage is reported as a number rather than assumed to be complete.
Gaps in the ladder create isobaric ambiguity. Where a fragment is missing, you observe a mass difference spanning two residues instead of one, and some two-residue sums are exactly equal to single residues:
| Ambiguity | Monoisotopic masses |
|---|---|
| Gly + Gly vs. Asn | 57.021 × 2 = 114.043 vs. 114.043 |
| Ala + Gly vs. Gln | 71.037 + 57.021 = 128.059 vs. 128.059 |
| Leu vs. Ile | identical — same elemental composition |
| Lys vs. Gln | 128.095 vs. 128.059 — ~0.036 Da apart |
BPC-157 illustrates two of these at once: an Ala-Gly pair at positions 12–13 that is isobaric with a single glutamine if the intervening fragment is missing, and a leucine at position 14 that standard CID cannot distinguish from isoleucine. Leu/Ile discrimination requires specialized fragmentation — EThcD, which follows electron-transfer dissociation with high-energy collision to generate diagnostic side-chain w-ions, or newer single-stage electron-activated dissociation approaches. These are characterization methods, not routine release tests.
For MOTS-c the point is sharper still: it carries both an isoleucine and a leucine, and its naturally occurring K14Q variant differs from wild type by roughly 0.036 Da — a difference we've flagged before as invisible on a unit-resolution intact-mass spectrum, and one that requires the fragment ladder to localize even at high resolution.
What MS/MS still cannot answer
Sequence confirmation is a large step up from intact mass. It is not the whole identity question.
Chirality remains invisible. Fragment ions from a D-residue and an L-residue have identical masses. This is not a minor caveat for a catalog containing SS-31 (D-Arg), FOXO4-DRI (all-D retro-inverso), and Dermorphin, whose single D-alanine is the residue that defines the molecule. Stereochemistry needs chiral analysis — a separate method, and not a line on a standard COA.
Disulfide connectivity needs a specific experiment. Cysteine pairing is settled by non-reduced peptide mapping: digest without reducing agent, so linked fragments stay linked, then identify the cross-linked species by LC-MS. The recognized hazard is that basic pH and elevated temperature during sample prep can themselves scramble disulfides, which is why the 2021–2025 methods literature is largely about suppressing that artifact — acidic digestion conditions, or cystamine plus low-concentration alkylating agent. For Oxytocin, a nonapeptide whose Cys1–Cys6 bridge is structural, and for IGF-1 LR3 with three disulfides, "purity 98%" says nothing about pairing.
Non-standard residues break automated interpretation. Database and search-engine workflows assume the twenty proteinogenic residues. Aib in Semaglutide, the acetyl group on Thymosin Alpha-1, a lipid side chain, a lactam bridge — each shifts fragment masses in ways a default search will not expect. Confirmation has to be targeted against the correctly annotated theoretical fragment list, or done de novo. The computational side of that is moving quickly; Bittremieux et al., Mass Spectrometry Reviews 2026;45(3):507–526 (online 29 Nov 2024) reviews the deep-learning methods that now dominate de novo sequencing.
And it says nothing about quantity. MS/MS is an identity technique. It does not measure net peptide content, water, counterion, or endotoxin.
Reading the documentation
Four checks, none of which require interpreting a spectrum:
- Is the named technique possible for this molecule? "Peptide mapping" on a proline-blocked heptapeptide is a template artifact. "MS/MS" on a 15-mer is appropriate.
- Is sequence coverage stated? A mapping or MS/MS claim without a coverage figure asserts that fragments were observed, not how much of the backbone they spanned.
- Is the theoretical fragment list built from the fully annotated sequence? Acetylation (+42 Da), amidation (−1 Da versus free acid), and non-standard residues must be in the calculation, or a correct product will appear to disagree.
- Does the claim match the failure mode that matters? Sequence order for a synthetic peptide; disulfide pairing for a cyclic or folded one; chirality for anything with a D-residue. A test only has value if it can detect something that molecule is capable of getting wrong.
More on documentation practice at /quality/, and compound-by-compound profiles at /library/.
FAQ
Does a full MS/MS ladder prove the compound is correct? It establishes sequence order to the extent of the observed coverage, which is a substantially stronger claim than a mass match. It does not establish stereochemistry, disulfide connectivity, or how much peptide is in the vial. Identity is layered, and no single method covers all of it.
Why would two labs report different sequence coverage for the same peptide? Fragmentation depends on charge state, collision energy, instrument type, and interpretation thresholds. As with HPLC purity and EC50, a coverage figure is comparable within a method and not across methods. The same proline-rich peptide can look well covered under one set of conditions and gapped under another.
Is MS/MS worth asking for on a short, simple peptide? It is most valuable where a wrong-order or truncated species would carry a plausible mass, which is a real risk for short sequences with repeated residues. For a molecule like TB-500, however, the first documentation question is not sequence order but which molecule is being sold — the heptapeptide or full-length thymosin beta-4 — and that one is answered by intact mass alone, as covered in BPC-157 vs TB-500. Match the test to the open question.
This article is educational and for the laboratory research community. Trulogic Labs products are sold for laboratory and research use only and are not for human consumption.